All experiments were done in accordance with Institutional Animal Care and Use Committee of Kyung Hee University and to minimize the number of animal experiments and suffering, we carried out the experiment with strict observance of the protocols and guidelines established by Kyung Hee University (KHUASP (SE)-16-059, 8 Aug 2016). Female Sprague-Dawley rats (10 weeks of age, 240~270 g, purchased from Daehan Biolink, introduced from Taconic Co., Albany, NY, USA) were housed under a 12:12 h (hour) light: dark cycle at an ambient temperature of 22℃. Water and rat chow were available ad libitum.

Stereotaxic surgery under chloral hydrate was performed as described [1,5,18]. Using coordinates relative to the bregma, stereotaxic injections of MPP⁺ (right medial forebrain bundle; A/P −3.6, M/L −2.0 D/V −7.7; MPP⁺, 7.4 µg in 2 µl phosphate-buffered saline, sigma) [1,5], or CNTFRαNAb antibody (right SN; A/P −5.5, M/L −2.2, D/V −7.7; R&D, AF-303-NA; 0.01 µg/µl, 0.2 µl/min, total 2µl) and respective control were done according to the atlas of Paxinos and Watson [18].

Capsaicin (1 mg/kg intraperitoneally; a single injection/day for 7 days, Sigma) [5,19] was injected at 4 or 9 week and 1 day post MPP⁺.

D-Amphetamine (5 mg/kg, intraperitoneally) was used to monitor ipsilateral rotations in rats with unilaterally lesioned nigrostriatal dopamine neurons [20]. The ipsilateral rotations were counted for 1 h at 3, 5, 8 or 10 weeks post MPP⁺.

As previously described [5,21,22,23], the total number of TH⁺ and Nissl⁺ cells was counted in the various animal groups using the optical fractionator method performed on a bright-field microscope (Olympus Optical, BX51) using Stereo Investigator software (MBF Bioscience). This unbiased stereological method of cell counting is not affected by either the reference volume (SNpc) or the size of the counted elements (neurons).

Optical densities of the TH⁺ striatal fiber were measured using Science Lab 2001 image Gauge (Fujifilm) [1].

Imaging data were analyzed in Image J (National Institutes of Health) as described recently [24]. Image J with co-localization plugin was used to quantify immunofluorescence and with color deconvolution plugin was used to quantify chromogenic signal intensity on image.

The SN area was rapidly removed and western blot analysis was performed as previously described [22,25]. The following primary antibodies and dilutions were used: rabbit TRPV1 (1:1000, Alomone labs), mouse anti-CNTF (1:1000, Millipore), mouse anti-GFAP (1:500, Sigma), mouse anti-tyrosine hydroxylase (TH, 1:1000, Millipore), rabbit anti-pTHser31 (1:1000, Millipore), and mouse anti-beta-actin (1:5000, Abcam). The following secondary antibodies and dilutions were used; horseradish peroxidase-conjugated anti- rabbit or mouse IgG (1:5000, Bethyl).

Animals were transcardially perfused, fixed and brain tissues (40 µm thick) and processed for immunohistochemical staining. In brief, sections were rinsed in phosphate-buffered saline (PBS) then incubated with the following primary antibodies: rabbit anti-TRPV1 (1:1000, Alomone labs) and rabbit anti-CNTF (1:200, Santa-Cruz), mouse anti-GFAP (1:500, mouse, Sigma) for astrocytes and rabbit anti-TH (1:2000, rabbit, Pel-Freez) for dopamine neurons. 3,3'-diaminobenzidine (DAB; Sigma) was used to visualize TH⁺ cell and fiber. 1% Cresyl violet (Sigma) solution was used for Nissl staining. The next day, tissues were rinsed and incubated with FITC-conjugated-anti-mouse goat IgG (1:400, Millipore) and/or Cy3-conjugated-anti-rabbit IgG (1:400, Millipore) for 1 h. The stained tissues were viewed using a confocal microscopy (LSM 700, Carl Zeiss) or were analyzed under a bright-field microscope (Olympus).

Levels of DA, DOPAC and HVA in STR were measured using the reversed-phase high performance liquid chromatography (HPLC,1260 Infinity system, Agilent Technologies, Santa Clara, CA) with electrochemical detector as described recently [26]. Dissected striatal tissues were homogenized with 0.1 M perchloric acid and 50 µM ascorbic acid per 10 mg brain tissue. Homogenate incubated on ice for 1h after sonication and centrifuged at 14000 rpm for 15 min, 4℃. the supernatants filtrated by 0.2 µm hydrophilic filter and injected into an auto-sampler at 4℃ (Waters 717 plus auto-sampler) and eluted through Sunfire C18 column (5 µm, 4.6×100 mm) with mobile phase for catecholamine analysis. The peaks of dopamine content were analyzed by ESA Coulochem III electrochemical detector (Dionex Corporation, Chelmsford, MA) and integrated using a commercially available program.

All values are expressed as mean ± standard error of the mean. Statistical significance (p<0.05 for all analysis) was assessed by One way ANOVA Newman-Keuls analyses and Student unpaired t-test using the Instat 3.05 software package (GraphPad Software, San Diego, CA, USA).

We first examined when MPP⁺ neurotoxicity could be complete. Immunohistochemical analysis demonstrated that MPP⁺ produces significant decreases in the number of TH⁺ and Nissl⁺ cells, assessed by stereology in the SNpc and density of TH⁺ fibers in the STR. When quantified and expressed as a percentage of cells on the MPP⁺-lesioned side compared to the PBS-treated control side, the number of TH⁺ and Nissl⁺ cells in the SNpc decreased by 68% and 63% at 2 weeks post MPP⁺, respectively (Fig. 1B, 1E, and 1J). Similar to those of 2 weeks, the number of TH⁺ and Nissl⁺ cells decreased by 78% and 65% at 10 weeks post MPP⁺, respectively (Fig. 1C, 1F, and 1J). In addition, the optical density of striatal fibers was significantly attenuated by 81% at 2 weeks and 82% at 10 weeks post MPP⁺ (Fig. 1H, 1I, and 1K). Taken together, the MPP⁺ model employed here showed a complete loss of nigrostriatal DA neurons at 2 weeks post MPP⁺ and no spontaneous recovery (restoration) of nigrostriatal DA neurons between 2 and 10 weeks post MPP⁺.

At 9 weeks post MPP⁺, rats which exhibited amphetamine-induced rotational asymmetry were randomly selected and treated with vehicle (Veh) as a control or CAP (i.p. 1 mg/kg; a continuous single injection per day for 7 days; Fig. 2A). At 10 weeks post MPP⁺, amphetamine-induced rotational asymmetry was significantly reduced in CAP-treated rats compared to Veh-treated rats (Fig. 2B). Next, we hypothesized that CAP effects on behavioral recovery are associated with neurorestoration on nigrostriatal DA neurons. Analysis by TH immunohistochemistry revealed that CAP treatment did not affect the number of TH⁺ cells, as assessed by stereology in the SNpc (Fig. 2C and 2D) and the density of TH⁺ fibers in the STR (Fig. 2C and 2E) in MPP⁺-lesioned rats compared to Veh-treated MPP⁺-lesioned control (Fig. 2C~E). Similar to TH immunohistochemical data, CAP treatment showed no substantial changes of Nissl⁺ cells compared to Veh treatment (Fig. 2D) in the SNpc of MPP⁺-lesioned rats compared to Veh-treated MPP⁺-lesioned control.

Our recent report demonstrates that CAP-induced behavioral recovery is attributable to endogenous production of CNTF, acting on CNTFRα expressed in DA neurons [1]. Accordingly, we tested the effects of CNTF and CNTFRα on behavioral recovery. CNTFRαNAb was unilaterally administered to inhibit CNTF actions in MPP⁺-lesioned SNpc at 9 weeks post MPP⁺ (Fig. 2A). CNTFRαNAb attenuated the effects of CAP on amphetamine-induced rotational asymmetry at 10 weeks post MPP⁺ (Fig. 2B), indicative of CNTF involvement.

Double immunofluorescence staining was performed on sections adjacent to those used for TH immunostaining in Fig. 1. Expression of TRPV1 in GFAP⁺ astrocytes was significantly increased in the SNpc at 2 weeks post MPP⁺, when MPP⁺ neurotoxicity was completed, and sustained up to 10 weeks post MPP⁺ compared with contralateral SNpc as a control (Fig. 3B and 3C). In a separate series of experiments on MPP⁺-lesioned rats, protein levels of TH, GFAP, and TRPV1 were measured at indicated time points (Fig. 3A). Western blot analysis demonstrated increases in GFAP expression with total TRPV1 levels unchanged in the SN, whereas TH levels were significantly decreased by 64% at 1 week, 81% at 2 weeks, and 80% at 10 weeks post MPP⁺, respectively, compared to the control (Fig. 3D and 3E). Taken together, these results indicate selective and prolonged expression of TRPV1 in GFAP⁺ astrocytes up to 10 weeks post MPP⁺.

As CNTF seems to be involved in CAP-induced behavioral recovery, and MPP⁺-lesion-induced increase in TRPV1 expression on astrocytes is sustained up to 10 weeks post MPP⁺. We wondered if CNTF could be endogenously produced by CAP-activated astrocytes through TRPV1 at 10 weeks post MPP⁺, leading to behavioral recovery. Immunohistochemical analysis revealed a significant increase in CNTF and CNTF expression in astrocytes in the SNpc of CAP-treated rats, compared to Veh-treated control at 10 weeks post MPP⁺ (Fig. 4A~C).

CNTF can stimulate TH activity via pTHser31 in the SN of MPP⁺-lesioned rats in vivo [1]. As expected, western blot analysis showed that, in the CAP-treated SN, increases in CNTF levels were accompanied by increases in pTHser31 levels compared to Veh-treated control SN, whereas total TH levels were unchanged at 10 weeks post MPP⁺ (Fig. 4D and 4E), reflecting no restoration of DA neurons in the SN (see Fig. 2). In addition, CNTFRαNAb attenuated CAP-induced increases in pTHser31 levels with unchanged total TH levels at 10 weeks post MPP⁺ (Fig. 4D and 4E), which might prevent CAP effects on amphetamine-induced rotational asymmetry (see Fig. 2B).

As amphetamine-induced rotational asymmetry is highly correlated with striatal levels of DA and its metabolites (DOPAC and HVA) [20], we wondered if increases in nigral TH activity by CNTF could elevate striatal DA biochemistry. Analysis by HPLC revealed that, in CAP-treated MPP⁺-lesioned rats, striatal levels of DA, DOPAC, and HVA were increased by 114% (p<0.01), 121% (p<0.01), and 365% (p<0.001), respectively, compared to Veh-treated MPP⁺-lesioned rats (Fig. 4F). Moreover, treatment with CNTFRαNAb partially reverted striatal levels of DA, DOPAC, and HVA (Fig. 4F). It is therefore likely that, compared to Veh-treated control, the increase in levels of DA and its metabolites by CAP treatment may be responsible for amphetamine-induced behavioral recovery in MPP⁺-lesioned rats.

The long-term effects of delayed CAP treatment on behavioral recovery were also determined after completion of MPP⁺ neurotoxicity on DA neurons (between 2 weeks and 10 weeks, see Fig. 1). Rats that exhibited amphetamine-induced rotational asymmetry at 3 weeks post MPP⁺ were randomly selected and treated Veh as a control or CAP (i.p. 1 mg/kg; a continuous single injection per day for 7 days), starting at 4 weeks post MPP⁺ (Fig. 5A). At 5 weeks post MPP⁺, CAP attenuated amphetamine-induced rotational asymmetry compared to Veh-treated control (Fig. 5B). Intriguingly, after termination of the first CAP treatment, behavioral recovery almost reverted to control levels (rotational score similar to Veh-treated MPP⁺-lesioned rats) at 8 weeks post MPP⁺ (4 weeks after the first CAP treatment) (Fig. 5B), indicating that CAP effects were temporal and persisted for no more than 3 weeks.

Then, all rats (n=6) that received the first CAP treatment (at 4 weeks post MPP⁺, see Fig. 5A) were randomly selected and divided into two groups: one (n=3) that received Veh as a control, and the other (n=3) that received the second CAP treatment (i.p. 1 mg/kg; a continuous single injection per day for 7 days), starting at 9 weeks post MPP⁺, respectively (Fig. 5A). At 10 weeks post MPP⁺, rats receiving CAP exhibited a significant reduction in amphetamine-induced rotational asymmetry compared to those receiving Veh as a control (Fig. 5B). Similar to the results shown in Fig. 2, immunohistochemical analysis revealed that CAP had no trophic effects on DA neurons in the SNpc and their fibers in the STR (Fig. 5C and 5D). As an additional control, rats that received Veh at 4 and 9 weeks post MPP⁺, without CAP treatment, showed no reduction in amphetamine-induced rotational asymmetry at 5 and 10 weeks post MPP⁺, respectively (Fig. 5B). Taken together, the present data suggest that behavioral recovery by delayed CAP treatment might be transient without neurorestoration and require continuous TRPV1 activation by CAP.